Gene expression profiles based Human cancer diseases classification

Abstract

Cancers are a large family of diseases that involve abnormal cell growth with the potential to spread to other parts of the body. A cancer disease in any of its forms represents a major cause of death worldwide. In cancer diagnosis, classification of different tumor types is of the greatest significance. Accuracy for prediction of various tumor types gives better treatment and minimization of toxicity on patients. Accordingly, creating methodologies that can effectively differentiate between cancer subtypes is essential. This paper presents a new methodology to classify Human cancer diseases based on the gene expression profiles. The proposed methodology combines both Information gain (IG) and Deep Genetic Algorithm (DGA). It first uses IG for feature selection, then uses Genetic Algorithm (GA) for feature reduction and finally uses Genetic Programming (GP) for cancer types' classification. The proposed system is evaluated by classifying cancer diseases in seven cancer datasets and the results are compared with most recent approaches.